summaryrefslogtreecommitdiffstats
path: root/src/plugins/dpdk
AgeCommit message (Collapse)AuthorFilesLines
2020-05-06crypto: fix coverity issuesFan Zhang1-5/+11
Fix coverity issues in crypto framework and cryptodev engine. Type: fix Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Change-Id: Ib261da0163c8182c803600db22c5a6dad5a19999
2020-05-05flow: explicitly convert RSS function types in dpdk_pluginChenmin Sun1-1/+34
explicitly convert RSS function types from vnet_rss_function_t to rte_eth_hash_function to avoid these two types go out of sync in the future... Type: fix Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Ic09f6bb7f2cfbcf7cc4d380e51554b7f2b7a3b90
2020-05-04fib: midchain adjacency optimisationsNeale Ranns1-14/+0
Type: improvement - inline some common encap fixup functions into the midchain rewrite node so we don't incur the cost of the virtual function call - change the copy 'guess' from ethernet_header (which will never happen) to an ip4 header - add adj-midchain-tx to multiarch sources - don't run adj-midchain-tx as a feature, instead put this node as the adj's next and at the end of the feature arc. - cache the feature arc config index (to save the cache miss going to fetch it) - don't check if features are enabled when taking the arc (since we know they are) the last two changes will also benefit normal adjacencies taking the arc (i.e. for NAT, ACLs, etc) for IPSec: - don't run esp_encrypt as a feature, instead when required insert this node into the adj's next and into the end of the feature arc. this implies that encrypt is always 'the last feature' run, which is symmetric with decrypt always being the first. - esp_encrpyt for tunnels has adj-midchain-tx as next node Change-Id: Ida0af56a704302cf2d7797ded5f118a781e8acb7 Signed-off-by: Neale Ranns <nranns@cisco.com>
2020-04-30flow: add l2tpv3oip flowChenmin Sun1-9/+45
This patch adds the l2tpv3oip type flow support Have tested on E810 with Intel iAVF driver Type: feature Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Icb5114b5f70dd7a63f681e7c6ac802fade8b8cf1
2020-04-30crypto: introduce async crypto infraFan Zhang5-9/+1435
Type: feature Signed-off-by: Damjan Marion <damarion@cisco.com> Signed-off-by: Filip Tehlar <ftehlar@cisco.com> Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Signed-off-by: Piotr Bronowski <piotrx.bronowski@intel.com> Signed-off-by: Dariusz Kazimierski <dariuszx.kazimierski@intel.com> Signed-off-by: Piotr Kleski <piotrx.kleski@intel.com> Change-Id: I4c3fcccf55c36842b7b48aed260fef2802b5c54b
2020-04-29misc: switch to clang-9Damjan Marion1-1/+1
Type: improvement Change-Id: Iebf77a63c0c19b130a3fbd26b5293304a9fed4c1 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-04-29flow: explicit rss function enum castFlorin Coras1-1/+1
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I8c0eb5b5b12ffac1ff5dc89cab10bdb8e4be4322
2020-04-28flow: add RSS supportChenmin Sun4-32/+80
This patch enables the RSS configuration through vnet/flow interface With this RSS feature, users can config the RSS functions for specific flows Currently, it supports: default, toeplitz and symmetric_toeplitz rss function, and ipv4-tcp/ipv4-udp/ipv6-tcp/ipv6-ucp flow types Users can use the following options to combine with above flow types for more specific hash input set selection: l3-src-only, l3-dst-only, l4-src-only, l4-dst-only Command line: test flow add dst-ip any proto udp rss function default rss types ipv4-tcp use l3-dst-only test flow add dst-ip any proto udp rss function toeplitz rss types ipv4-udp use l4-src-only test flow add dst-ip any proto udp rss function symmetric_toeplitz rss types ipv6-udp use l3-src-only and l3-dst-only Type: feature Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I213efc76dc8af37f2f63605884f353e05b0f5d2a
2020-04-27vlib: deprecate i2c and cjDave Barach4-4/+0
i2c follows its only use case - the original 82599 driver - into extras/deprecated. cj is/was an emergency debug tool unused in several years. Move to extras/deprecated/vlib Type: refactor Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: Ib55b65373f62630db295c562974bd8f2456c3107
2020-04-22misc: fix coverity warningsDave Barach1-2/+3
Type: fix Ticket: VPP-1837 Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: I9ec87d2293d8f92c3e488a0f61083cf815ac496c
2020-04-22devices: Adding ConnectX-6 DX Device ID for validationAmir Zeidner1-2/+7
Type: fix Change-Id: I3c7ebbe91e7dffe9fd6851e5334fe920f2187cf0 Signed-off-by: Amir Zeidner <amirzei@mellanox.com>
2020-04-17dpdk: fix udp-encap for esp in transport modeAlexander Chernavin2-10/+9
Now UDP encapsulation doesn't work in transport mode because: - the encrypt node misses filling of UDP header and it gets sent with all zeros; - the decrypt node misses filling of new IP header and it contains garbage data. With this commit, fill UDP header during encryption and fill IP header during decryption. Change-Id: I87a7bd594f0e312b16d3e5eb19e568b4e3164d36 Type: fix Signed-off-by: Alexander Chernavin <achernavin@netgate.com>
2020-04-02dpdk: VM VHOST tests with rxq>1 failingSteven Luong1-0/+1
DPDK recently added a check in the virtio driver to make sure that rxmode->mq_mode == ETH_MQ_RX_NONE. We were passing ETH_MQ_RX_RSS and the device initialization was not accepted. The reason for the change in DPDK was that there is no controls (algorithm, redirection table, hash function). So they thought ETH_MQ_RX_NONE was the best choice for the value of mq_mode. Type: fix Ticket: VPP-1853 Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: Ifa0fc4206cedc56a851f94f6434a2a7500bbd419
2020-03-27dpdk: enable DPDK iAVF PMDChenmin Sun3-0/+23
This patch adds support for the DPDK iAVF PMD Type: feature Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I7bb0f621774e4c55b9b7309462e6591ce1b88fb6
2020-03-13dpdk-ipsec: Fix for multipoint IPSECNeale Ranns1-4/+4
Type: fix fetch the sa_index from the correct location Change-Id: I351035ee0226c47585995ff9122320fd5c73ec53 Signed-off-by: Neale Ranns <nranns@cisco.com>
2020-03-12dpdk: tx/rx burst function description refactorChenmin Sun1-6/+31
DPDK provides two new APIs to retrieve information about the Tx/Rx packet burst mode: rte_eth_tx_burst_mode_get rte_eth_rx_burst_mode_get This patch leverages these two APIs to describe the tx/rx mode. Currently, Intel X710/E810 and Mellanox Mlx5 support the new APIs. For NICs that don't support the new APIs, still use the original way to print their tx/rx function name Type: refactor Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Ibe47f5debe3b3f17f462fbf9834394e22845cc08
2020-03-04flow: add vlan tagged types for IPv4/IPv6 5-tuple flowsChenmin Sun1-3/+6
Add new flow type IP4_N_TUPLE_TAGGED and IP6_N_TUPLE_TAGGED for vlan tag sensitive flows The original IP4_N_TUPLE and IP6_N_TUPLE will not match VLAN anymore Type: feature Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Ie511e9a64126440fe81f29665a56ca060061662d
2020-02-20dpdk: rx checksum offloadFlorin Coras1-0/+8
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I5a3e3a41dcc60c0d9b291e51bb112e7701f73050
2020-02-14dpdk: TSO does not work for Cisco VICSteven Luong1-0/+11
While TSO is supported for Intel NIC, Cisco VIC does not work. The problem is due to txmode offloads is not properly set for the Cisco VIC when enable-tcp-udp-checksum is configured. Type: fix Ticket: VPP-1838 Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: I72c41db9b327ed8d08ef70d74e8cc6206d4a102f
2020-02-14dpdk: fix flow(with mark action) deletion crash issueChenmin Sun1-1/+1
Type: fix this patch fixes mark flow deletion crash issue, see below test flow add src-ip any proto udp src-port 111 dst-port 222 mark 100 test flow enable index 0 1/1 test flow disable index 0 1/1 test flow enable index 0 1/1 test flow disable index 0 1/1 -> [crash] This is because the code resets a wrong vector in flow lookup entry recycle logic. See function dpdk_flow_ops_fn(). Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I2b0a1e531931ab25541d672d88da18dc2289f1ce
2020-02-13dpdk: Add iova-mode to startupVladimir Ratnikov1-1/+2
In some cases of using vfio-pci driver, it should be required to enable pa or va mode. Without it, rtl_eal_init unable to allocate memory required. Debugging told that iova-mode pa/va could help. And it helps. This patch allows to pass iova-mode to vpp startup.conf Type: feature Change-Id: I36b87f5d3d141891b37cda2c306d50433954a34a Signed-off-by: Vladimir Ratnikov <vratnikov@netgate.com>
2020-02-05dpdk: use port_id as interface name suffix for representorsHaggai Eran1-1/+7
Type: feature Representor devices include a port ID as part of their switch_info struct, and it is helpful to use that in the interface name. Signed-off-by: Haggai Eran <haggai.eran@gmail.com> Change-Id: Id24627e7daf857f8b0e8ace2f592c098678081c7
2020-02-05dpdk: output switch informationHaggai Eran1-0/+21
Type: feature Output DPDK switch information to allow finding out which DPDK ports are associated with which DPDK representor ports. Signed-off-by: Haggai Eran <haggai.eran@gmail.com> Change-Id: I612cbd5a97e04787eca13423f53c7283d5945e37
2020-01-30misc: deprecate dpdk hqosDamjan Marion11-3355/+0
Not in functional state for a long time ... Type: refactor Change-Id: I2cc1525a6d49518cbc94faf6afbf0d2d0d515f56 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-01-22ipsec: re-enable DPDK IPSec for tunnel decap/encap (VPP-1823)Neale Ranns4-5/+27
Type: fix Change-Id: Iff9b1960b122f7d326efc37770b4ae3e81eb3122 Signed-off-by: Neale Ranns <nranns@cisco.com>
2020-01-14dpdk: enforce max tx retriesBenoît Ganne1-0/+1
n_retry was never decremented and so never enforced. Type: fix Change-Id: I71d60a72c156286f7e5b82b1c77a723361317c69 Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-12-17flow: add ethernet flowChenmin Sun1-2/+68
Type: feature This patch adds the ethernet type flow support Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I6de163d04c6472df60a5f296fb428efd000267c4
2019-12-17dpdk: add devargs supportChenmin Sun3-12/+29
Type: feature This patch adds the devargs support for dpdk device The devargs are used as hardware-specific init args for dpdk devices please refer to the nic guides under $(DPDK_DIR)/doc/guides/nics/$(NIC_DRIVER).rst Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Id380d04720090bb66afe5ce09d664e5e248b8eb9
2019-12-06dpdk: use explicit types in apiOle Troan1-3/+4
Type: fix Signed-off-by: Ole Troan <ot@cisco.com> Change-Id: Iff7e995499e422003147342ff14de483d807b3f3
2019-12-04gso: remove the interface countMohsin Kazmi1-2/+0
Type: refactor Change-Id: I51405b9d09fb6fb03d08569369fdd4e11c647908 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-11-26flow: fix several crash issuesChenmin Sun1-2/+0
Type: fix This patch fixes crash issues(marked in brackets) in the below test cases test flow enable index 0 1/1 -> [crash] test flow disable index 0 1/1 -> [crash] test flow add src-ip 192.168.8.8 proto udp redirect-to-queue 8 test flow enable index 0 1/1 test flow disable index 0 1/1 - [crash] test flow add src-ip 192.168.8.8 proto udp redirect-to-queue 8 test flow enable index 0 1/1 test flow enable index 0 1/2 test flow del index 0 -> [crash] Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I84bc6faa3d93a2cab4c82e8a876a8b1067257b62
2019-11-20flow: Add GTP supportChenmin Sun1-28/+260
Type: feature Adding: VNET_FLOW_TYPE_IP4_GTPC VNET_FLOW_TYPE_IP4_GTPU VNET_FLOW_TYPE_IP4_GTPU_IP4 VNET_FLOW_TYPE_IP4_GTPU_IP6 VNET_FLOW_TYPE_IP6_GTPC VNET_FLOW_TYPE_IP6_GTPU VNET_FLOW_TYPE_IP6_GTPU_IP4 VNET_FLOW_TYPE_IP6_GTPU_IP6 in this patch Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I4ad53895b5ac0771432bb039b8c79e48e3c19f25
2019-11-19dpdk: fix non-NULL terminated stringBenoît Ganne1-1/+1
Type: fix Change-Id: Ic221cd4fcad89aece71239ed96152bf0311f3286 Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-11-14dpdk: Enable bnxt PMDAndy Gospodarek3-1/+20
This adds support to compile the bnxt PMD that supports Broadcom's 10/25/40/50/100/200 Gbps NICs. Tested with modified DPDK driver on x86_64 and aarch64 targets that is pending upstream acceptance, but should compile just fine today. Type: feature Signed-off-by: Andy Gospodarek <gospo@broadcom.com> Change-Id: I735a991c4cae4fa77e5605094facea54b781a1db
2019-11-07dpdk: ipsec gcm fixesChristian Hopps3-25/+14
- Fix AAD initialization. With use-esn the aad data consists of the SPI and the 64-bit sequence number in big-endian order. Fix the u32 swapped code. - Remove salt-reinitialization. The GCM code seems inspired by the GCM RFCs recommendations on IKE keydata and how to produce a salt value (create an extra 4 octets of keying material). This is not IKE code though and the SA already holds the configured salt value which this code is blowing away. Use the configured value instead. Type: fix Change-Id: I5e75518aa7c1d91037bb24b2a40fe4fc90bdfdb0 Signed-off-by: Christian Hopps <chopps@labn.net>
2019-11-06dpdk: conditionally revert rte_vfio_dma_map patchDamjan Marion1-0/+5
DPDK have bug which will be fixed in 19.11. Type: fix Change-Id: I6c0058928e5991d61b3c5fcba706f35e6886b0f2 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-11-01dpdk: fix tso not properly check the 'enable-tcp-udp-checksum' option issueChenmin Sun1-16/+15
Type: fix Fix tso did not properly check the 'enable-tcp-udp-checksum' option issue Add description of 'tso' and 'enable-tcp-udp-checksum' in startup.conf Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Id659067a9fa9e1db6c3f8dc533a2e90351b86831
2019-10-30dpdk: use local loggerPaul Vinciguerra1-5/+5
Type: fix Change-Id: I44922f70aef6a3c53f0f56c6d0656502c8fd69b2 Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
2019-10-29dpdk: replace deprecated VFIO API invocationDamjan Marion1-1/+2
Type: refactor Change-Id: I1712729b3f5d4ab659f51a2d3bf798db16b50820 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-10-28vlib: don't use vector for keeping buffer indices in the poolDamjan Marion1-11/+7
Type: refactor Change-Id: I72221b97d7e0bf5c93e20bbda4473ca67bfcdeb4 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-10-16dpdk: add function to add/del extra MAC addrsMatthew Smith1-0/+24
Type: feature Add a callback to install or remove an additional MAC address on a DPDK-managed device. Note that some PMDs don't have support for this so YMMV. Change-Id: I7b0cb3fb0af0d66fbdb7b894a712897f889520a5 Signed-off-by: Matthew Smith <mgsmith@netgate.com>
2019-10-16flow: Add 'drop' and 'redirect-to-queue' actions supportChenmin Sun1-3/+20
Type: feature Add 'drop' and 'redirect-to-queue' support in 'test-flow' command and DPDK plugin Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I567bb77cb401c9bd1309ecabe802fe9de88c746b
2019-10-03dpdk: remove api boilerplateOle Troan6-236/+11
Type: refactor Signed-off-by: Ole Troan <ot@cisco.com> Change-Id: I2cba1e2912902b092348dddcf9cbcba231c75995
2019-09-30ipsec: allow null cipher with dpdk espChristian E. Hopps1-1/+2
The NULL cipher is a (valid) non-AEAD choice for ESP encrypt path. Allow it. Type: fix Signed-off-by: Christian E. Hopps <chopps@chopps.org> Change-Id: I6d8b66223a0ffb0952c2dd6fa898a8a2289fef7a
2019-09-24vlib: add flag to explicitelly mark nodes which can init per-node packet traceDamjan Marion2-0/+2
Type: feature Change-Id: I913f08383ee1c24d610c3d2aac07cef402570e2c Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-09-16dpdk: initialize rte_mbuf during mempool dequeueDamjan Marion1-1/+44
In some cases it may happen that buffer is allocated by DPDK, and freed by VPP native code. In such cases dpdk metadata is not reset, so we need to do that during mempool dequeue. Template approach is taken to reduce cost of that operation. Type: fix Fixes: 910d369 Change-Id: Ic239007cfc8fbceb965021c56963cda9d53f63be Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-09-12dpdk: fix null-termination of stringsElias Rudberg1-0/+3
Fix problem with some strings not being null-terminated, by using the vec_terminate_c_string macro in two places. The problem was found using AddressSanitizer. (Also make sure indentation is OK for those changes.) Ticket: VPP-1772 Type: fix Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net> Change-Id: Ib7826e3c322e58b649e2d7f6053786da618a5e9e
2019-09-11dpdk: apply dual loop unrolling in DPDK TXLijian.Zhang1-0/+57
Too many prefetches within loop unrollings induce bottleneck and performance degradation on some CPUs which have less cache line fill buffers, e.g, Arm Cortex-A72. Apply dual loop unrolling and tune prefetches manually to remove hot-spot with prefetch instructions, to get throughput improvement. It brings about 1% throughput improvement and saves 8% clocks with the target node on Cortex-A72. Type: feature Change-Id: If3a64a04a77e90cd0240bc4d1186dbb09dac7df0 Signed-off-by: Lijian Zhang <Lijian.Zhang@arm.com>
2019-09-05dpdk: fix extended statsFilip Tehlar1-1/+1
Extended stats are not displayed due to incorrect condition. Type: fix Change-Id: Ie04664e6274137462dce832bf7ee06204cd77be5 Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
2019-08-30dpdk: enable checksum offload for Intel SRIOV NIC driversJohn Lo1-0/+8
Type: fix Signed-off-by: John Lo <loj@cisco.com> Change-Id: I3ee59106ba45164a4ee3788bf4dcf5bf4c2dc1c2